The Use of Inductive Logic Programming for the Developement of the Statistical Software Tool 
CORA (system description)

Iris Pigeot und Ursula Robers Ludwig-Maximilians-Universität München, Universität Dortmund, 
Germany

Abstract 

This paper presents the knowledge acquisition process for the system CORA, which provides a sta-
tistical analysis of stratified case-control data and additionally supports the researcher at the 
different steps of this analysis. In addition to the general information given by the help system,
the advice tool recommends an adequate statistical method. This recommendation depends on the 
actual data to be analysed. The underlying rule base containing the recommendations was discovered
by inductive logic programming (ILP) methods.